ISSN (Online) 2395-2717

International Journal of Engineering Research in Electrical and Electronic Engineering

(IJEREEE)
Volume 11 Issue 01 January 2024

Power Output Optimization of a Non-Uniformly

Illuminated Photovoltaic System Using Dynamic
Switching Reconfiguration
[1]
Julito B. Añora Jr., [2] Jitter Josef Johann L. Quilonia, [3] Arch Mechkael Y. Sumapig
[1]
Department of Electrical Engineering CIT - University, Cebu Philippines
[2]
Department of Electrical Engineering CIT - University, Cebu Philippines
[3]
Department of Electrical Engineering CIT - University, Cebu Philippines
[1]
[email protected], [2] [email protected], [3] [email protected]
ORCID ID 0000-0003-4479-9291

Abstract— The solar Photovoltaic (PV) systems’ efficiency and performance are greatly affected because of shading caused by cloud
cover and shadows from trees. At this condition, PV power output decrease because the shaded solar panels act as load and thus dissipate
power generated by the PV systems. This study will focus on the power output optimization of the PV at shading conditions using a new
switching strategy: “groups of operation”. Using this strategy, PV panels belonging to the mth and nth column with the greatest number
of shaded panels are reconfigured dynamically to form various array configurations. MATHLAB/Simulink model is developed to
simulate the power outputs of the PV systems given the conditions of shading and malfunctioning. Simulation results show an increase
of power output after reconfiguration given different possibilities of panels randomly shaded; 44.25% for one shaded panel, 12% for two
shaded panels at different row, 179.47% for 2 shaded panels at the same row, 43.03% for three panels at different row, and 141.92% for
three shaded panels at same row. The simulation further shows that for more than three shaded panels in the 3x3 array, an optimized
power output inasmuch as 86.36% to 151% is recorded after switching reconfiguration. HOBOWARE Experimental results of the 3x3
setup showed that the switching strategy optimizes the power output inasmuch as 13.48% to 110.42%.

Index Terms— Non-uniform solar insolation, PV systems, Solar irradiance, Uniform illumination.

polluting and unsustainable, and are characterized by

I. THE PROBLEM AND ITS SETTINGS inequity in consumption and access [6] [7]. These situations
will eventually create economic and political chaos among
A. Introduction
nations once supply of energy will be depleted and used up.
Energy is an essential component of developments that With world economic development, conventional energy
drives the economy of every nation and the world. Daily sources have become increasingly unable to meet the worlds
needs such as food and housing, transportation, delivery of ‘growing demand for energy. This imminent threat to energy
health and sanitation, education, businesses, among others security must be approached thoroughly by maximizing the
require energy to support human activities. renewable energy sources in the form of wind, tidal,
The global demand of energy has skyrocketed [1], and geothermal, hydro, and solar energy. In Visayas and
primary energy consumption grew by 1.8% in 2012, well Mindanao, almost two-thirds of base load energy is provided
below the last ten years average of 2.6%. Regionally, the by renewable sources accounting 69.52% and 62%,
greatest contribution to growth took place in emerging respectively [8].
economies, in particular China and India, which explains The solar has proven to be an effective source of renewable
90% of the increase in global energy consumption in the energy. In 2004, the first and largest grid-connected solar
OCDE fell by 1.2% [2]. In Asia, including the Philippines, photovoltaic power plant in Southeast Asia was set- up in
the weight on total primary energy consumption will increase Cagayan de Oro. On the third year of its operations, annual
from 40% in 2010 to 45% in 2050 [3]. energy generation was 10% higher than expected. An
A continuous growth can be noticed in total energy expansion into Misamis Oriental of up to 30 MW is planned
demand, while hydrocarbons and fossil fuels continue to have for Mindanao [8].
a significant share in the total structure. Oil consumption The PV systems market in the Philippines is so vast. The
increased by nearly 900,000 barrels per day(bl/day) or nearly country is archipelagic; and where logistics is one of the
1%, and coal consumption grew by 2.5% far from the problems, conventional power distribution in an inter-island
historical average of 4.4% [2]. In 2030, it is forecasted that grid requires big investments. In small islands, power is
fossil fuels will still be available, but there is a geographical driven by diesel generators, but its sustainability is threatened
imbalance between the energy supply and demand that will by the increasing costs and the price volatility of oil in the
eventually lead to the exhaustion of fossil fuels [4] [5]. world market. The solar energy, therefore, offers renewable
Globally, current patterns of energy consumption are

5
 ISSN (Online) 2395-2717
International Journal of Engineering Research in Electrical and Electronic Engineering
(IJEREEE)
Volume 11 Issue 01 January 2024
advantage in power generation over conventional sources in ii. Design a hardware switching reconfiguration
the remote islands and far-flung areas. Apart from being where the PV panels do not remain in a static
clean and emission-free, solar panels are portable and connection with respect to each other but can
scalable making it more viable in the sun’s energy extraction. dynamically adjust their connection to form
However, the potential of this energy generation is various array configurations to yield maximum
challenged by non-uniform illumination due to shadows cast energy given different scenarios e.g.
by trees, poles, clouds, bird litters and the growing urban shadowing, PV malfunction, etc.
environments such as buildings and residential units. The iii. Characterize the improvements in the output
non-uniform illumination affects the power output of the PV power of the PV System once switching
system regardless of its scale and magnitude. reconfiguration algorithm is implemented.
In [9], the effects of partial shading have significantly
reduced the real power generated by a PV system. The II. METHODOLOGY
shading causes problems such as irregular P-V characteristics This study aims to develop an algorithm and
(multiple peaks) which makes power optimization difficult methodologies that will improve the performance and
by the conventional power extraction techniques [10]. The efficiency of the PV system during shading conditions using
shading also causes damage to solar cell arrays when local switching reconfiguration of the solar panels. Specifically, it
hot spots are concentrated with an excessive flow of current, will utilize a switching strategy where the Photovoltaic
thus reducing the life service of the PV systems. This panels do not remain in a static connection with respect to
proposed research therefore addresses the problems on partial each other but can dynamically adjust their connection to
shading on PV systems by developing algorithms and form various array configurations to yield maximum energy
methodologies to optimize power output, and allow given different scenarios e.g. shadowing, PV malfunction,
self-reconfiguration of solar panels whenever it is partially etc. Hence, this chapter will present research methodology
shaded or totally damage. that will advance the implementation of the project towards
The Philippines is geared towards achieving energy its goals.
security through renewables with a target of 45.15% RE The research outlines three major procedures as follows:
increase by 2017. Emerging technologies such as the solar i. Conceptualization of the research project
will increasingly contribute to the energy mix leading up to including the system overview, designs of the
2020 [11]. sensors, microcontroller, and actuators.
B. Statement of the Problem ii. Implementation of the research design based on
the detailed and technical specifications. A
The PV systems’ efficiency and performance are greatly
laboratory experimental setup is also outlined.
affected because of shading caused by cloud cover and
iii. Evaluation and analysis of the research outputs
shadows from trees. First, the shaded cells can get reverse
and data for the systems switching modeling. It
biased and it consume instead of generating power resulting
also includes iteration of the procedures to verify
in loss of total output power [12]. Second, the power losses in
the validity and adaptability of the systems model
the individual shaded cells would result in local heating and
that relates to the objective of this paper.
increase the temperature affecting surrounding cells. The
increase in temperature creates thermal stress on the entire
module and cause hot spots and local defects which
potentially result in the failure of the entire array [12] [13].
Third, if one full cell is completely shaded, the voltage of that
module will drop to half of its unshaded value in order to
protect itself [14] [15]. In a case of a hybrid Grid - PV system,
shading causes power output intermittency which leads to
grid instability. Existing solutions require additional panels
to electrically “replace” the shaded panels. However, there is
no solution implemented in industry scale. More importantly,
the life expectancy and service of the PV system will be
shortened. In this regard, this study aims to do the following:
i. Develop an algorithm, methodologies and
simulation studies of solar panels and Fig. 1 – Research Methodology
reconfiguration using MATLAB/Simulink.

6
 ISSN (Online) 2395-2717
International Journal of Engineering Research in Electrical and Electronic Engineering
(IJEREEE)
Volume 11 Issue 01 January 2024
A. Conceptualization
1) PV Reconfiguration System Overview
Accordingly, the prototype will comprise a (3 x 3)
mini-solar array, microcontroller, and a set of actuators such
as the Solid-State Relays (SSR). The solar arrays are initially
arranged in a series-parallel connection as shown in Figure
2.6.

Fig. 3 PV Panels Conditioning Circuit Using

Opto-coupler.

3) Algorithm and Methodologies Development

a) MATLAB/Simulink Simulation of One Solar Panel
One of the specific objectives in this study is to develop an
algorithm, methodologies and simulation studies of solar
Fig. 2 PV Reconfiguration System Overview panels and reconfiguration using MATLAB/Simulink. Doing
so would require the simulation of a single diode model in
Figure 2.2 shows the overall system overview of the Figure 2.1. Six-solar panels connected in series are also
proposed methodology. The input variable to the 3x3 solar considered for simulation to see the effects of the variability
panels is the solar irradiance ranging from 0 W/m2 (such a of the solar irradiance to the PV system. Two different PV
case of PV malfunctioning) to 1000 W/m2 (effective solar specifications (see Table 1 and Table 2) will separately be
irradiance at optimal condition). An optocoupler simulated using the MATLAB model in order to verify the
conditioning circuit is designed to sense the output voltage of adaptability of the simulation model at different PV scale. A
each panel at any conditions. The analog values of voltage comparative analysis of two simulations will be presented
will be transmitted as input to the Arduino AT2560 including the behaviors of solar panels at shading.
microcontroller. When the microcontroller receives the For the 3x3 experimental setup, a separate
signal from the conditioning circuit, it will undergo signal MATLAB/Simulink model using the Simscape toolbox is
processing to come up with a digital output (DO) for decision designed to characterize the switching reconfiguration
algorithms of switching reconfigurations. The actuators (i.e. methods of the panels. For each simulation, the behaviors of
Solid-State Relay) are then directed by the microcontroller the PV and VI curve characteristics will be recorded to see
(through the DO) to dynamically arrange the solar panels at what configurations optimize the power output of the PV
shading conditions. The input to the microcontroller is an system at shading conditions. When a switching strategy
analog voltage ranging from 0 - 5V. combination is seen to optimize the power output of the PV
2) PV Panels Optocoupler Conditioning Circuit to the system, it will be used to develop the switching algorithms in
Microcontroller the hardware prototype.
As mentioned, each solar panel in the 3x3 experimental Table 1: Photovoltaic Specifications 1 For Simulation
setup will have a signal conditioning circuit for the voltage
input to the microcontroller. To eliminate the independent
ground loops in the system, an optocoupler is used to separate
the input signals to the output signals. An optocoupler is a
device or component that transfers electrical signals by
means of light. A current limiting resistor is designed to limit
the current flowing through the Light Emitting Diode (LED).
A 5 kΩ resistor is used as a current limiter to match the
impedance of the PV panels to that of the optocoupler. A
pull-down resistor is connected in the emitter side of the
optocoupler. The purpose is to convert the generated current
signal to voltage signals as an input to the Microcontroller
ranging from 0 Volts to 5 Volts, respectively.

7
 ISSN (Online) 2395-2717
International Journal of Engineering Research in Electrical and Electronic Engineering
(IJEREEE)
Volume 11 Issue 01 January 2024
Table 2: Photovoltaic Specification 2 For Simulation ii. Alternative Routing Switch Method (ARSM)
The hardware switching reconfiguration requires the
performance of the Solid-State Relays being able to trigger
dynamically at times during solar PV reconfiguration. These
switches are dynamic enough to be expanded for an mth x nth
panels subjected for reconfiguration.
For the purpose of this study, the dynamic route switch
method is introduced. The principle behind this method is
b) Hardware Switching Reconfiguration Algorithm associated with the concept of road routing in the main
The overall hardware switching reconfiguration setup is highways in the city. As can be noted, a public vehicle which
shown in Figure 2.2. The experimental set-up will comprise 3 has a definite route will surely pass in and out certain
x 3 mini solar panels, a terminal box, a microcontroller, intersections in the road. For this analysis, the concept of
switch box and an optocoupler voltage conditioning circuit switch routing is applied with an assumption that when a
module. The DC - DC solid state relay is being used for this current enters one electrical node, the same current will leave
application with an input voltage range of 3-32 Volts DC. at that same node.
This input voltage to the SSR falls within the output voltage Each panel in the 3x3 experimental setup is numbered P1
capacity of the Arduino AT2560 microcontroller. for PV (1,1), P2 for PV (1,2), P3 for PV (1,3), P4 for PV
The switching reconfiguration is implemented using the (2,1), P5 for PV (2,2), P6 for PV (2,3), P7 for PV (3,1), P8 for
“groups of operation” principle of the panels (see Section 5 (3,2) and P9 for PV (3,3), respectively. Likewise, in the
for the discussions). This group operation is singly used in implementation of the software algorithms, numbers 1-9
this study to determine the panels’ subject for reconfiguration generally referred to panels PV (1, 1) - PV (3, 3). The
based on the least number of shaded panels in the rows or designed notation is utilized as serial I.D’s for the
columns of the original PV connection. implementation of the algorithms in the Arduino Mega
AT2560 microcontroller.
i. Algorithm Rules Based on the Proposed
Each panel in the setup is routed with the other panels for
Methodology: Groups of Operation
the dynamic reconfiguration to happen. In between these
The switching reconfiguration of the panels is routes is SSR’s labeled S1- S36. The algorithm is so designed
implemented using the “group of operation principle.” Using that only the switches within the route will be triggered thus
this methodology, the panels belonging to the mth row or nth avoiding the possibilities of short circuiting within the PV
columns with the greatest number of shaded panels is system. An added feature in the algorithm is the identification
assumed to operate within the “group of operation.” of the PV panels subjected for shading and malfunctioning. A
Accordingly, the rules defined for this methodology are display of the serial monitor shows the status of the PV
outlined: system, including the readings of each PV panel, the number
1. For a 3x3 experimental setup, a panel is said to be in a of shaded panels and its corresponding location.
“group of operation” if it belongs to the greatest
number of shaded panels in a row or column. The Table 3: Alternative Route Switch Table
“group of operation” is either a horizontal or vertical
group of operation. Only the panels identified within
the groups of operation will reconfigure itself
independently with the panels belonging to the other
groups.
2. In a case of a row or a column where no shaded panels
are sensed, it shall be left independently alone with the
rest of the panels belonging in the “group of
operation”. These panels will not be reconfigured, but
rather be a point of connection with the rest of the
panels in the “groups of operation”.
3. In reference to Rule 1, the panels’ in the “group of
operation” will be connected in the following
combinations: series, series - parallel, parallel - series,
and parallel, respectively. In a case of more than two
groups arise in the system; one group of operation will
be reconfigured with the other groups.

8
 ISSN (Online) 2395-2717
International Journal of Engineering Research in Electrical and Electronic Engineering
(IJEREEE)
Volume 11 Issue 01 January 2024
iii. Solar PV Reconfiguration for Shading Strategy disregarded since both the module position in a row and the
The implementation of the algorithms in the Arduino row position in the single string are irrelevant with respect to
Mega AT2560 microcontroller is predefined based on the the delivered DC output Power [16].
results of the MATLAB/Simulink simulation. The
discussions in Chapter 4 show different reconfiguration
results of the partially shaded panels in all the conditions of
PV malfunctioning. In each of the shading conditions, the
power output possible for all configurations is graphically
outlined to distinguish what combinations optimizes the
output of the PV system. This configuration will be noted and
implemented in the switching algorithms of the hardware
prototype. In the reconfiguration process, there are rules to
follow to ensure uniformity of the algorithm.
4) Solar Switching Reconfiguration and Possible Fig. 6 Irrelevant configurations with respect to the
Combinations delivered DC Power [16]

In this regard there are only 280 configurations of the 362

880 ones can deliver different values of maximum DC Power
at the PV output. At any mth row and nth column, the general
formula of finding the maximum relevant combination is:

5) Solar Array Groups of Operation

Fig. 4 3x3 Experimental Setup with Optocouplers used in

the Voltage Sensing to the Microcontroller

Fig. 7 Solar Array Horizontal Group of Operation

This study proposes switching reconfigurations of the solar

panels within its group of operation. A “group of operation”
is a term solely used in this study which means boundaries by
which the solar panels possibly operate. The groups can be
series or parallel, a diagonal or a combination depending on
which optimizes the power output. Conversely, the mth row
and an nth column of the solar array can be considered a
“group of operation” and the panels belonging to this mth and
Fig. 5 3x3 Solar Arrays nth row, respectively, is said to operate within the group.
The principle of this methodology is that, whenever the
The solar switching strategy will do the reconfiguration of solar panels are shaded, the remaining panels within the
the PV system once the system detects considerable change group shall reconfigure itself independently with the rest of
in the solar irradiance. For a 3x3 solar array there are N = the groups. The localization of the reconfiguration is meant to
(m.n)! = (3.3)! = 362,880 combinations. However, as shown simplify the control algorithms of the switching strategy by
in Figure 2.7, some of the possible configurations can be means of grouping the panels to operate within its

9
 ISSN (Online) 2395-2717
International Journal of Engineering Research in Electrical and Electronic Engineering
(IJEREEE)
Volume 11 Issue 01 January 2024
boundaries. Figure 2.7 shows a horizontal group of be isolated in the array. The remaining four panels: PV (1,2),
operations 1, 2 and 3 of the 3x3 solar arrays. Whenever PV PV (1,3), PV (2,1) and PV (3,1) will be reconfigured
(1,1) is severely shaded and/or damaged, the remaining accordingly leaving the panels PV (2,1), PV (2,3), PV (3,2)
panels in the group: PV (1,2) and PV (1,3) will reconfigure and PV (3,3) static with respect to the original circuit
itself independently with the rest of the panels while the connections as shown in Section 3, Figure 2.3.
shaded panel PV (1,1) is isolated in the entire system. Only
the panels operating in its group of operation will reconfigure
itself. Other panels in the mth row and or nth column shall
work and reconfigure independently with the other solar
panels.
In [18], the shaded panels that triggered the switching
reconfiguration can be dealt with in two ways. First, if the
shading in the panel is significant so that the solar irradiance
approximates 85% reduction from the nominal irradiance of
1000 W/m2 and/or 0 W/m2 such as in a case of a damage PV
system, the shaded panel will be bypassed temporarily until
conditions to make it operate in the strings are met. Second, Fig. 8 Horizontal and Vertical Groups of Operations of
in a case of solar irradiance reduction greater than 15% of the One Shade Panel
nominal irradiance, it can be connected in parallel with the
other shaded cells. Its terminal voltage shall be ensured that it As stated, the simulation of each partial shading case is
will meet the nominal operating voltage of the other strings to carried out by assuming that the PV panels are completely
prevent these shaded panels to behave as loads rather than as isolated when it is partially shaded. This is a valid
a source in the system. This can be carried out by employing methodology since the main objective of the research is to not
a DC-DC boost converter in the output terminals of the PV just optimize the power output of the PV system at shading
systems. In the implementation of the automatic switching condition but also protect the PV solar cells from thermal
reconfiguration, the threshold voltage can be set at values stress and overheating (that also leads to the cracking of the
higher than the 85% reduction to ensure that PV cells can solar cells in due consideration of the outdoor weather
operate at a safer value than the proposed value in [18]. conditions) during partial shading.
The reduction of the solar irradiance values can be found In the 3x3 panels, the solar array reconfiguration is (m.n)
using a solar irradiance sensor/pyranometer. However, in the (3x3)! = (9!) = 362,880. As emphasized in Fig. 2.7, there
absence of the pyranometer sensor, the solar irradiance can are a total of 280 configurations that will yield relevant
be approximated using the output values of voltage and configurations. The 280 combinations are a result when all
current using the formula: panels in the 3x3 solar arrays are reconfigured relative to
each other. However, in the proposed reconfiguration
method, the panels that will be reconfigured belonged to the
specific groups of operation.
Where Gij is the estimated irradiance, and Vij the
measured current and voltage, respectively, and a, Io and
nVT is a set of parameters which can be evaluated from the
values of the short circuit current, the open circuit voltage,
and the maximum power point of the PV module given in the
data sheets of the manufacturer, respectively [16].
6) Effective Combination for Shaded Panels
In relation to Section 5 (see Figure 2.7), the solar arrays
are grouped into zones which generally are referred as
“groups of operation”. In this principle, the solar panels can
be reconfigured within its groups of operation independently
from the other arrays whenever a shaded panel or panels are
sensed. In general, the groups of operation can be horizontal
Fig. 9 Irrelevant Configuration of the Panel Using the
or vertical or a combination of both. The intersection of both
Principle of Groups of Operation
the horizontal and vertical groups of operation in Figure 2.8
(i.e., PV (1,1)) is assumed to be the shaded panel that needs to

10
 ISSN (Online) 2395-2717
International Journal of Engineering Research in Electrical and Electronic Engineering
(IJEREEE)
Volume 11 Issue 01 January 2024
For a one completely shade panel, the maximum PV output terminal to the DC male and female jacks.
combination of the panels is 4! = 24. In these 24
combinations, there are irrelevant configurations that need
not be simulated, leaving the effected combination to be at 14
configurations.
Fig. 2.9 shows one PV reconfiguration when PV (1, 1) is
severely shaded. PV (1, 2) is connected in parallel with the
series combination of PV (1, 3), PV (2, 1) and PV (3, 1),
respectively. Assuming each panel outputs 5 V at maximum
solar irradiance, PV (1, 2) branch has 10 V disparities as
compared to the other branch. The output voltage in each
branch does not satisfy the requirements of connecting one
electrical unit to the others in parallel. In the simulation of PV
reconfiguration in MATLAB/Simulink, the combination that
does not meet the requirements of connecting one panel in
parallel to the others need not be simulated and disregarded. Fig. 10 Output terminal
As a result, when the number of panels within the group of
operation accounts to 4 panels, there would be 14 effective
3) Terminal Box
combinations that will yield distinct power for one case of
solar panel severely shaded. The terminal box will be used to accommodate all nine
The proposed solar switching methodology has solar panel outputs into one enclosure. Its features include the
narrowed down the number list of possible switching 9-DC Power Female Plug Jack where the male jack plug of
reconfiguration. The tedious process of selecting a the PV output terminal will be inserted and served as the
configuration that will yield distinct power output through input to the conditioning circuit prior to inputs to the
complex switching algorithms will be minimized while microcontroller. Figure 11 shows the terminal box layout.
optimizing the power output of the PV system at shading
conditions. The panels are then reconfigured manually in the
MATLAB/Simulink model (see Results and Discussions). In
each of the configurations, the PV Curve and the VI Curve
characteristics are then gathered and compared which yields
the optimum power. The combination that yields the
optimized power output will be noted to be the configurations
of interest in the implementation of the hardware switching
reconfiguration design using the microcontroller and the
Solid-State Relays. Fig. 11 Terminal Box(left); optocoupler conditioning
cirucit (right)
a) Implementation
1) Hardware Components and Specifications
b) Evaluation
In the course of the implementation of the hardware
In the evaluation of the data gathered in the
switching strategy, several electrical components and
implementation phase, the designs presented in Section 2.1
accessories are evaluated as to how to carry out the
including the types of microcontrollers, value of the resistors
implementation of the 3x3 experimental setup prototypes.
in the voltage divider, among others will be iterated or change
These accessories will be in a plug-and-play setup were
if deemed necessary in order to meet the objectives presented
gathering of data would be easier.
in Section 1.2.
2) PV Output terminals
The photovoltaic module output terminals will be III. RESULTS AND DISCUSSION
configured and designed that it can be connected easily to the A. Simulation and Experimental Setup Validation
terminal box. It will be a plug-and-play set-up. The output
The 3x3 laboratory experimental setup and the switching
terminal conductors of the PV system will each have a male
algorithm have an actual light intensity of 350 W/m2. The
plug jack. Likewise, the terminal box will have a mounted
light intensity is measured using the pyranometer sensor
DC female plug jack in order to connect the PV output
embedded in the prototype box housing the 3x3 solar panels.
terminals directly. Figure 2.6 shows the connections of the

11
 ISSN (Online) 2395-2717
International Journal of Engineering Research in Electrical and Electronic Engineering
(IJEREEE)
Volume 11 Issue 01 January 2024
The same light intensity is entered into the the MATLAB/Simulink simulation in order to come up with
MATLAB/Simulink model simulation. To enter the same a comparison of the percent change of power when switching
light intensity both in the simulation and the experimental algorithm is implemented.
setup is intended to compare and characterize the
improvements of the power output of the PV system once
dynamic switching reconfiguration is implemented. The
experimental data for voltage and current is gathered using
the HOBOWARE voltage and current transducers. Figure 12
shows the experimental setup of the solar reconfiguration.

Fig. 14 Solid State Relay Housing Module and the 3x3

solar arrays

Note that the hardware prototype algorithm is based on the

predefined results on the MATLAB/Simulink model.
Therefore, the comparison of results for the percent change of
power comprises the following: MATLAB/Simulink from
before configuration of each case of shading,
MATLAB/Simulink after configuration of each case of
shading, the experimental result in each case of shading using
Fig. 12 Experimental Setup Using the HOBOWARE
the HOBOWARE data. Fig. 14 shows the solid-state relay
housing module and the 3x3 solar array housing.

Fig. 15 Solid State Relay Housing Module and the 3x3

Fig. 13 Experimental Setup Housing for Solar Panels and solar Arrays
Pyranometer Sensor Prototype.

The sensors being utilized are the HOBOWARE current

and voltage transducers. Fig. 13 shows the prototype box of
the experimental setup housing the solar panels and the
pyranometer sensor. The laptop is used in the data gathering
of the parameters needed in the analysis.
The data are gathered before and after the reconfiguration
process. An adjustable light source comprising 3-100W
incandescent bulb, controlled by a dimmer switch is used in Fig. 16 Experimental Results of the 3x3 solar PV setup.
the experimental setup. First, the light intensity is measured Current and Voltage for all panels at 350 W/m2(left).
using the pyranometer sensor. Once the numerical value of Reconfiguration of the system after 1 panel is simulated for
the intensity is determined, the same intensity value is used in shading (right)

12
 ISSN (Online) 2395-2717
International Journal of Engineering Research in Electrical and Electronic Engineering
(IJEREEE)
Volume 11 Issue 01 January 2024
Except in the case of one shaded panel, the data before 1) Simulation and Experimental Validation: 1 shaded
configuration in the experimental result can possibly be panel
gathered because of the default configuration in the switching
algorithm. As can be seen in Fig. 16, the results for a normal
3x3 array have 12.706 Volts and current 6.5411 mA across
the connected load to 17.134 Volts and 7.42 mA after
reconfiguration. This shows that for one shaded panel, the
reconfiguration optimizes the power output of the PV system.

Fig. 18 Configuration Results comparison for MATLAB

and Experimental Results: 1 shaded panel

Fig. 18 shows the MATLAB/Simulink simulation and the

Experimental results when reconfiguration happens at a time
Fig. 17 Transitional data from 1 shaded panel
when 1 panel is randomly shaded. In the graph, 87.87% is
reconfiguration (left) to 2 shaded panel reconfiguration(right) recorded in an increase of power when implemented in
MATLAB/Simulink. However, the experimental results
In Fig. 17, the HOBOWARE data shows the voltage and
show only 52% increase of power after reconfiguration. This
current outputs increasing after the reconfiguration of the is higher than the power output of the PV system when
solar panels when 1 panel is shaded to 2 panels shaded at
reconfiguration is not implemented in MATLAB/Simulink
different row. The voltage and current during the
model for 1 shaded panel.
reconfiguration when 2 panels are shaded are 28.914 Volts
and 9.78 mA from the 17.59 Volts and 7.51 mA, respectively.
2) Simulation and Experimental Validation: 2 shaded
panels at different row
B. Characterization of Power Output:
MATLAB/Simulink simulation results Vs. the
Experimental Results Using the HOBOWARE
As previously noted, the 3x3 solar array experimental
setup has 5kΩ, connected across its terminals. The
HOBOWARE voltage and current transducers are connected
across and in series this load to determine the actual values of
voltage and current when reconfiguration happens at any case
when shading occurs in the solar panels. The light intensity
value is determined using the pyranometer sensor embedded
in the prototype box of the solar panels. When the intensity is
determined, the same value is entered to the
MATLAB/Simulation and the model is simulated again with
the same load connected across its terminals. Experimental
results showed that for a laboratory experimental setup, the
average light intensity that the pyranometer sensor can Fig. 19 Configuration Results comparison for MATLAB
measure is at 350 W/m2. and Experimental Results: 2 shaded panel at different row

Fig, 19 shows the configuration results comparison for

MATLAB and the experiments using the HOBOWARE. As
can be seen, MATLAB simulation registers 289.24%
increase of power after reconfiguration while experimental

13
 ISSN (Online) 2395-2717
International Journal of Engineering Research in Electrical and Electronic Engineering
(IJEREEE)
Volume 11 Issue 01 January 2024
setup using the HOBOWARE indicates 110.42% increase of 5) Simulation and Experimental Validation: 4 shaded
power. The graph further shows that after reconfiguration, panels at different row and column
both MATLAB/Simulink and HOBOWARE register an
increase of power by more than a 100%.
3) Simulation and Experimental Validation: 2 shaded
panels at same row

Fig. 22 Configuration Results comparison for MATLAB

and Experimental Results: 4 panels randomly shaded.

In Fig. 22, four panels are randomly shaded. Using

MATLAB/Simulink simulation, the reconfiguration of the
remaining panels’ registers an increase of power at 210.11%
while 40.99% is registered using the HOBOWARE in the
Fig. 20 Configuration Results comparison for MATLAB experimental setup. The reconfiguration in both the
and Experimental Results: 2 shaded panel at different row experimental and simulation shows a significant increase of
power for the PV system.
When two panels are shaded in the same row,
6) Simulation and Experimental Validation: 3 shade
MATLAB/Simulink shows an increase of power by 262.32%
panels at same column
after reconfiguration as shown in Fig. 20. Experimental
results show an increase of 38.894%. The experimental When panels in the same column are randomly shaded, the
results for this reconfiguration proved to increase the power MATLAB/Simulink registers 343.92% increase of power
of the PV system after reconfiguration. while 87.93% is registered for the experimental setup using
HOBOWARE.
4) Simulation and Experimental Validation: 3 shaded
panels at different row and column

Fig. 23 Configuration Results comparison for MATLAB

and Experimental Results: 3 panels at the same column
Fig. 21 Configuration Results comparison for MATLAB
and Experimental Results: 3 shaded panel at different row
C. Groups of Operation Concepts and the Relevant
When three panels in the 3x3 experimental setup are Research
shaded randomly, 124.21% is recorded in increase of power 1) Irradiance Equalization Method
using the MATLAB/Simulink model. The reconfiguration,
The research in [16], proposes an irradiance
however, registers a 13.488% increase of power in the
equalization method in the reconfiguration of the solar panels
experimental setup using the HOBOWARE.
during shading conditions. The method is applicable only to a

14
 ISSN (Online) 2395-2717
International Journal of Engineering Research in Electrical and Electronic Engineering
(IJEREEE)
Volume 11 Issue 01 January 2024
system that is connected in Total – Cross – Tied (TCT) IV. CONCLUSION
configuration. Since the reconfiguration of the solar panels The reconfiguration of the solar panels in the 3x3
depends on the extent of the irradiance received by the experimental setup has proven to increase the power output
modules, the number of the switches position in between the of the PV system at times during partial shading and PV
panels are significantly large in numbers. Since the algorithm malfunctioning. The “Groups of Operation” switching
continues to match the irradiance of each row, the current and strategy not only proves to increase the power output of the
voltage sensors connected for each panel would be large in PV system but also simplifies the complexity of the algorithm
number as the system is increased to an mth by nth array. The associated by the increasing number of SSR’s for an increase
complexity of the algorithm and the increasing numbers of mth x nth array. The implementation of this switching strategy
sensors and transducers add up to the cost of the PV system is made even simpler with the introduction of Alternative
[19]. And although, there is reconfiguration that takes place Routing Switch Method (ARSM).
for the solar panels during irradiance equalization, it might In the simulation of partial shading and PV
not be the optimal topology. malfunctioning, the 1st MATLAB/Simulink model proves to
2) Adaptive Banking accurately emulate the power output. The method of
connecting six solar panels in series to characterize the
In research [17], the proposed solar reconfiguration
effects of partial shading is a valid assumption noting that the
composes two parts: the fixed part and the adaptive part. The
results of the simulation also coincides with the
fixed part is connected in Total Cross Tied configuration
manufacturers rating and specifications of the PV panels. The
while the adaptive part comprises of individual panels
2nd MATLAB/Simulink model is also dynamic as different
arranged in an additional (n +1)th column to the mth by nth
PV ratings can be entered in the dialog box window so
array. When the power output of the PV array in the fixed
designed for this application. The characterization of the
part goes down, the modules from the adaptive bank are
power output is shown by gathering the actual voltage and
connected in parallel to modules in the fixed part. This is
current data at different shading conditions using the
accomplished by employing a switching matrix constructed
HOBOWARE current and voltage transducers. A 52%
using relays or electrical switches. Compared to the
increase of power for one shaded panel, 110.42% for 2
irradiance equalization method, it requires a switching matrix
shaded panels at different row, 38.89% for 2 panels at same
of smaller size as it relocates only the module in the adaptive
row, 13.49% for 3 panels at same row, 87.93% at same
bank rather than all the modules in the array. However, it still
column, respectively, are recorded in the experimental setup.
uses a non-optimal fixed array topology and needs an
In general, this research has addressed two things. First,
additional set of modules allocated as the adaptive bank. The
it is proven that the power output can be optimized using
additional modules increase the cost of the system. [19] The
dynamic switching reconfiguration inasmuch as 44% to
power output as claimed by the authors is recorded to be at
179.47%. Second, the ARSM addresses the problems of the
30% in a typical experiment [17].
increasing cost for the SSR in any switching reconfiguration.
3) Groups of Operation (GOO) Randomly shaded panels can now be simulated without
The proposed switching reconfiguration noted as computing all possible configurations when two or more
“groups of operation” configures the solar panels by panels are damaged as ARSM provides a distinct route
arranging the panels in groups either horizontal or vertical. algorithm that also prevents short circuiting of the PV system
The GOO switching strategy proves to increase the power during reconfiguration.
output of the 3x3 solar panels since 179.47%. The grouping
of the solar panels in the proposed strategy is meant to RECOMMENDATIONS
simplify the switching algorithm. To achieve the If the concept of this research is to be implemented in
optimization of the switches, an alternative switching larger scale, the specifications of the DC-DC SSR’s has to be
strategy is derived to complement the groups of operation so carefully designed to meet the output voltage of the PV
method - the Alternative Routing Switch Method (ARSM). panels. The power losses of the system have to be analyzed
By having this algorithm, the switching strategy is eased by further to ensure that the PV’s power output is delivered to
triggering specific solid-state relays distinctly as to avoid the the connected loads in a more optimal manner. The transient
short circuiting of the panels during the actual effects during each time the switches are reconfigured should
reconfiguration process. The complexity of the algorithm is also be analyzed noting that sudden voltage fluctuations and
addressed by both the GOO strategy and the ARSM. The time delays can affect the voltage stability should a PV
number of SSRs in the ARSM method is also reduced thereby system is connected to the grid.
minimizing the cost of the system.

15
 ISSN (Online) 2395-2717
International Journal of Engineering Research in Electrical and Electronic Engineering
(IJEREEE)
Volume 11 Issue 01 January 2024
REFERENCES
[1] C. J. M. a. M. Cheek, Clean Energy Nation, 2012.
[2] E. A. Pelegry, "Energy: Current Situation and Trends. From
Global to Local," Orkestra-Instituto Vasco de Competitividad,
2014.
[3] E. A. Pelegry, "Energy Scenarios and Trends Section".
[4] K. Kurokawa, Energy from Desert, ISBN 1 902916417, 2003.
[5] D. D. Nguyen, "Modeling and Reconfiguration of solar
Photovoltaic Arrays under Non-uniform Shadow Conditions".
[6] N. a. NJOMO, "Analyse du systeme energetique camerounais
dans une perspective de developpemnet soutenable," p. p. 588,
March-April 2009.
[7] M. E. a. D. NJOMO, "Mathematical Modeling and Digital
Simulation of PV Solar Panel using MATLAB Software,"
International Journal of Emeergung Technology and
Advanced Engineering, Vols. vol. 3, pp, no. ISSN 2250-2459,
2013.
[8] L. L. A. C. H. O. J. A. P. R. d. M. M. Dia, "The Philippine
Energy Roadmap to 2020," GReenPeace SouthEast Asia,
2011.
[9] W. J. T. Unruh, "Limits on Cloud-Induced Fluctuation in
Photovoltaic Generation," IEEE Trasnsactions on Energy
Conversion, Vols. vol. 5, no. 1, pp. pp. 8-14, 1990.
[10] E. T. S. A. B. P. P. Sharma, "REconfiguration Strategy for
Optimization of Solar Photovoltaic Array Under Non-uniform
Illumination Condtions," in Department of Electrical
Engineering, IIT Bombay Mumbai,India.
[11] G. a. D. o Energy, "The Philippine Energy Revolution
ROadmap 2020," 2011.
[12] P. O. Singh, "Modeling of Photovoltaic Arrays under Shading
Patterns with Reconfigurable Switching and Bypass Diodes".
[13] D. S. a. U. R. Hanitsch, "Shading Effects on Output Power of
Grid Connected Photovoltaic Generator Systems," in
REvised. Energ.Ren.: Power Engineering, Berlin Germany,
2001, pp. pp. 93- 99.
[14] P. P. H. S. G. Pachpande, "Studying the Effect on Solar Panel
Using Matlab," International Journal of Science and Applied
Information Technology," vol. vol. 1, no. no. 2, 2012.
[15] N. Y. S. H. K. H. O. a. K. K. Naka, "Simulation of I-V
Characteristics of PV module with shaded PV Cells," Solar
Energy Matter.Solar Cells, vol. vol. 90, no. no. 3, p. p. 329,
2003..
[16] F. G.-G. R. P.-L. M. R.-L. A. C.-R. G. Velasco-Quesada,
"Electrical PV Array Reconfiguration Strategy for Energy
Extraction Improvement on Grid- Connected PV Systems,"
IEEE Transactions on Industrial Electronics, vol. vol. 56, no.
no. 11, 2009.
[17] P. S. E. T. S. D. a. V. A. B. Patnaik, ""Reconfiguration
Strategy for Optimization of Solar Photovoltaic Array Under
Non-Uniform Illumination Conditions," in Department of
Electrical Engineering, IIT Bombay Mumbai, India..
[18] S. B. V. K. C. T. A. S. M. B. H. Braun, "Topology
reconfiguration for optimization of photovoltaic array output,"
Sustainable Energy, Grids and Networks 6, 2016.
[19] B. L. D. Nguyen, "An Adaptive Solar Photovoltaic Array
Using Model-Based Reconfiguration Algorithm," ii.

16